An Experiment on Using Genetic Programming to Generate Multiple Software Variants
by R. Feldt
PDF
Software fault tolerance schemes often employ multiple software variants developed to meet the same specification. If the variants fail independently of each other, they can be combined to give high levels of reliability. While design diversity is a means to develop these variants, it has been questioned because it increases development costs and because reliability gains are limited by common-mode failures. We propose the use of genetic programming to generate multiple software variants by varying parameters to the genetic programming algorithm. We have developed an environment to generate programs for a controller in an aircraft arrestment system. Eighty programs have been developed and tested on 10000 test cases. The experimental data shows that failure diversity is achieved but for the top performing programs its levels are limited.
Bibtex
@TechReport{Feldt1998GPSWVariants,
author = "Robert Feldt",
title = {An Experiment on Using Genetic Programming to Generate Multiple Software Variants},
year = "1998",
institution = "Dept. of Computer Engineering, Chalmers University of Technology, Gothenburg",
number = "Tech. report 98-13",
keywords = "Automated Programming; Genetic Programming; Controlled Experiment",
url = "http://www.cse.chalmers.se/~feldt/publications/feldt_1998_experiment_gp_variants.html",
}